From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dinh Nguyen Subject: Re: SoCFPGA ethernet broken Date: Thu, 15 Oct 2015 21:32:18 -0500 Message-ID: References: <561FF9E2.30102@opensource.altera.com> <56200687.9040903@gmail.com> <562005AD.8020903@opensource.altera.com> <56200BD7.8020505@gmail.com> <56200E15.9080603@caviumnetworks.com> <56201158.8040806@opensource.altera.com> <56201AE7.2020103@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Cc: David Daney , "David S. Miller" , , , "linux-kernel@vger.kernel.org" To: Florian Fainelli Return-path: In-Reply-To: <56201AE7.2020103@gmail.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: netdev.vger.kernel.org On Thu, 15 Oct 2015, Florian Fainelli wrote: > On 15/10/15 13:49, Dinh Nguyen wrote: > >> > >> Does this text change with and without the 8b63ec1837fa patch? > > > > No, this text does not change with/without the 8b63ec1837fa patch. > > Could you instrument mdiobus_scan(), get_phy_device() and > phy_device_create/register to see if the parent is NULL, non-NULL? > Yes, I can do that. > So far, I cannot see what is wrong with David's changes, quite the > contrary, and if there was something wrong with the PHY device creation, > it should not get you that far. > > You have not answered my previous question though, do you have PHY > fixups registered for that ID? By fixups, do you mean the skew values that are in the device tree? Those are the only fixups that I have the PHY. Another debugging point, the SoCFPGA board has a Micrel ksz9021 PHY attached to the ethernet port. What I'm seeing is that with 8b63ec1837fa patch, when the call to ksz9021_config_init() is made both of_node and dev->parent->of_node are NULL, without the patch the dev->parent->of_node is a valid pointer. Thus the skew values get programmed to the phy. BR, Dinh